explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-11-14 13:47:13 +03:00
Code Issues Packages Projects Releases Wiki Activity
14,070 Commits 68 Branches 175 Tags 279 MiB
compat/lemmatizer-factory-defaults
Commit Graph

136 Commits

Author SHA1 Message Date
Matthew Honnibal
ffc371350a
Avoid assuming encode.get_dim('nO') is set in tok2vec (#6800) 2021-01-24 14:37:33 +11:00
Sofie Van Landeghem
c8761b0e6e
rewrite Maxout layer as separate layers to avoid shape inference trouble (#6760) 2021-01-19 07:37:17 +08:00
Adriane Boyd
26c34ab8b0
Fix parser resizing for cupy (#6758) 2021-01-18 20:43:15 +01:00
Matthew Honnibal
c2a18e4fa3 Update textcat ensemble model 2021-01-19 02:53:02 +11:00
Ines Montani
a203e3dbb8 Support spacy-legacy via the registry 2021-01-15 21:42:40 +11:00
Ines Montani
b0b743597c Tidy up and auto-format 2021-01-15 11:57:36 +11:00
Sofie Van Landeghem
75d9019343
Fix types of Tok2Vec encoding architectures (#6442) 
* fix TorchBiLSTMEncoder documentation

* ensure the types of the encoding Tok2vec layers are correct

* update references from v1 to v2 for the new architectures
 2021-01-07 16:39:27 +11:00
Sofie Van Landeghem
3983bc6b1e
Fix Transformer width in TextCatEnsemble (#6431) 
* add convenience method to determine tok2vec width in a model

* fix transformer tok2vec dimensions in TextCatEnsemble architecture

* init function should not be nested to avoid pickle issues
 2021-01-06 12:44:04 +01:00
Ines Montani
991669c934 Tidy up and auto-format 2021-01-05 13:41:53 +11:00
Sofie Van Landeghem
282a3b49ea
Fix parser resizing when there is no upper layer (#6460) 
* allow resizing of the parser model even when upper=False

* update from spacy.TransitionBasedParser.v1 to v2

* bugfix
 2020-12-18 18:56:57 +08:00
Sofie Van Landeghem
cfc72c2995
Bugfix multi-label textcat reproducibility (#6481) 
* add test for multi-label textcat reproducibility

* remove positive_label

* fix lengths dtype

* fix comments

* remove comment that we should not have forgotten :-)
 2020-12-09 06:29:15 +08:00
Sofie Van Landeghem
de108ed3e8
Add specific error when StaticVectors can't read the vectors data (#6450) 2020-12-09 06:16:07 +08:00
Sofie Van Landeghem
f98a04434a
pretrain architectures (#6451) 
* define new architectures for the pretraining objective

* add loss function as attr of the omdel

* cleanup

* cleanup

* shorten name

* fix typo

* remove unused error
 2020-12-08 14:41:03 +08:00
Sofie Van Landeghem
a0c899a0ff
Fix textcat + transformer architecture (#6371) 
* add pooling to textcat TransformerListener

* maybe_get_dim in case it's null
 2020-11-10 20:14:47 +08:00
Sofie Van Landeghem
75a202ce65
TextCat updates and fixes (#6263) 
* small fix in example imports

* throw error when train_corpus or dev_corpus is not a string

* small fix in custom logger example

* limit macro_auc to labels with 2 annotations

* fix typo

* also create parents of output_dir if need be

* update documentation of textcat scores

* refactor TextCatEnsemble

* fix tests for new AUC definition

* bump to 3.0.0a42

* update docs

* rename to spacy.TextCatEnsemble.v2

* spacy.TextCatEnsemble.v1 in legacy

* cleanup

* small fix

* update to 3.0.0rc2

* fix import that got lost in merge

* cursed IDE

* fix two typos
 2020-10-18 14:50:41 +02:00
Sofie Van Landeghem
f8a1c1afd6
avoid dropout at runtime (#6247) 2020-10-13 14:39:59 +02:00
svlandeg
40276fd3be update NEL docs after latest refactor 2020-10-12 11:41:27 +02:00
svlandeg
08cb085f6c Merge remote-tracking branch 'upstream/develop' into fix/various 2020-10-09 17:01:27 +02:00
svlandeg
040c7c0541 fix get_dim calls in build_simple_cnn_text_classifier 2020-10-09 15:40:58 +02:00
svlandeg
853edace37 fix MultiHashEmbed example in documentation 2020-10-09 14:11:06 +02:00
Adriane Boyd
39aabf50ab Also rename to include_static_vectors in CharEmbed 2020-10-09 11:54:48 +02:00
Matthew Honnibal
cfb9770a94
Fix empty input into StaticVectors layer (#6211) 
* Add test for empty doc(s)

* Fix empty check in staticvectors

* Remove xfail

* Update spacy/ml/staticvectors.py
 2020-10-06 14:15:41 +02:00
Ines Montani
1a554bdcb1 Update docs and docstring [ci skip] 2020-10-05 21:55:27 +02:00
Ines Montani
9614e53b02 Tidy up and auto-format 2020-10-05 21:55:18 +02:00
Matthew Honnibal
e50047f1c5 Check lengths match 2020-10-05 20:02:45 +02:00
Matthew Honnibal
cdd2b79b6d Remove deprecated MultiHashEmbed 2020-10-05 19:58:18 +02:00
Matthew Honnibal
6dcc4a0ba6 Simplify MultiHashEmbed signature 2020-10-05 19:57:45 +02:00
Matthew Honnibal
eb9ba61517 Format 2020-10-05 15:29:49 +02:00
Matthew Honnibal
8ec79ad3fa Allow configuration of MultiHashEmbed features 
Update arguments to MultiHashEmbed layer so that the attributes can be
controlled. A kind of tricky scheme is used to allow optional
specification of the rows. I think it's an okay balance between
flexibility and convenience.
 2020-10-05 15:22:00 +02:00
Ines Montani
bcd52e5486 Tidy up errors and warnings 2020-10-04 11:16:31 +02:00
Ines Montani
3bc3c05fcc Tidy up and auto-format 2020-10-03 17:20:18 +02:00
svlandeg
02247cccaf Merge remote-tracking branch 'upstream/develop' into feature/small-fixes 2020-10-02 20:48:11 +02:00
Matthew Honnibal
6965cdf16d Fix comment 2020-10-02 17:26:21 +02:00
Ines Montani
af282ae732 Fix import 2020-10-02 01:12:34 +02:00
Ines Montani
e59ecb12c0 Auto-format 2020-10-02 01:12:30 +02:00
Matthew Honnibal
75a1569908 Merge 2020-10-01 23:07:53 +02:00
Matthew Honnibal
300e5a9928
Avoid relying on NORM in default v3 models (#6176) 
* Allow CharacterEmbed to specify feature

* Default to LOWER in character embed

* Update tok2vec

* Use LOWER, not NORM
 2020-10-01 23:05:55 +02:00
Matthew Honnibal
b854bca15c Default to LOWER in character embed 2020-10-01 22:17:58 +02:00
Matthew Honnibal
684a77870b Allow CharacterEmbed to specify feature 2020-10-01 22:17:26 +02:00
Sofie Van Landeghem
a22215f427
Add FeatureExtractor from Thinc (#6170) 
* move featureextractor from Thinc

* Update website/docs/api/architectures.md

Co-authored-by: Ines Montani <ines@ines.io>

* Update website/docs/api/architectures.md

Co-authored-by: Ines Montani <ines@ines.io>

Co-authored-by: Ines Montani <ines@ines.io>
 2020-10-01 16:22:48 +02:00
svlandeg
5121972930 add types of Tok2Vec embedding layers 2020-10-01 09:20:09 +02:00
svlandeg
5a9fdbc8ad state_type as Literal 2020-09-23 17:32:14 +02:00
svlandeg
25b34bba94 throw custom error when state_type is invalid 2020-09-23 16:57:14 +02:00
svlandeg
dd2292793f 'parser' instead of 'deps' for state_type 2020-09-23 16:53:49 +02:00
svlandeg
6c85fab316 state_type and extra_state_tokens instead of nr_feature_tokens 2020-09-23 13:35:09 +02:00
Ines Montani
1114219ae3 Tidy up and auto-format 2020-09-21 10:59:07 +02:00
Adriane Boyd
f3db3f6fe0
Add vectors option to CharacterEmbed (#6069) 
* Add vectors option to CharacterEmbed

* Update spacy/pipeline/morphologizer.pyx

* Adjust default morphologizer config

Co-authored-by: Matthew Honnibal <honnibal+gh@gmail.com>
 2020-09-16 17:45:04 +02:00
Ines Montani
1955aaaa20
Merge pull request #6045 from svlandeg/feature/more-layers-docs [ci skip] 2020-09-09 21:46:40 +02:00
Sofie Van Landeghem
cb66ea7400
Remove simple_ner code (#6041) 
* remove simple_ner code

* remove unused _biluo and _iob files
 2020-09-09 16:11:27 +02:00
svlandeg
39aa740777 Merge remote-tracking branch 'upstream/develop' into feature/more-layers-docs 2020-09-09 11:59:34 +02:00